JPH034352A - System for controlling non-routine data processing program - Google Patents

Abstract

PURPOSE:To execute a high efficiency processing by inferring data contents with a main program from the keyword of a data group, referring to an index table, calling an optimum instruction table and giving a control to a corresponding processing module. CONSTITUTION:The keyword of the data group inputted from an OS is inferred by a main program 1, an index table group 3 to describe the optimum processing module is referred to, and the value of a branch destination is decided. By the value, the optimum instruction table is called from a branch instruction table 4, the processing module corresponding to the call result is selected from a module group 2, and the control is given to the module. By the constitution, the processing module close to processing conditions is selected, and the non- routine processing program is highly efficiently processed.

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial Application Field) The present invention relates to a computer that processes the contents of atypical data randomly generated from a plurality of terminal devices, etc. using a processing module corresponding to the data. This paper relates to control methods for non-routine data processing programs in systems.

(Prior Art) In a computer system that processes the content of non-standard data randomly generated from a plurality of terminal devices, etc. using a processing program corresponding to the data, keywords are selected from the non-standard data and further processing is performed. In order to arrive at the processing program to be processed, a judgment is made based on a list of conditional comparison instructions, and this is done by branching to the corresponding processing program.

When input data, such as document information that flows between an unspecified number of people such as an e-mail system, is created in a non-standard manner, when checking and processing this data, the algorithm must be created each time, and multiple processing methods are required. If similar processing is to be performed in a program, it is necessary to change the program sequentially based on the programmer's experience.

At this time, program maintenance could always affect the entire program.

(Problems to be Solved by the Invention) That is, since checks are always performed using the same condition comparison logic, there are problems in that loss time occurs and the dedicated memory increases due to similar logic.

Furthermore, there is a problem in that program maintainability is impaired because work is required mainly to review the logic due to changes in conditions and the like.

The present invention was made to solve these problems, and it searches for data using human-generated atypical data as keywords, and uses the rough inference results to efficiently create a processing program that corresponds to the data content. The purpose is to provide a non-standard data processing program control method for passing data.

[Structure of the Invention] (Means for Solving the Problems) The atypical data processing program control method of the present invention that achieves the above object processes the contents of atypical data randomly generated from a plurality of terminal devices, etc. In a computer system that processes using modules, the main program infers the content of the data from the keywords of the data group manually, accesses the inte x table that describes the optimal processing module as a result, and uses the information in the inte x table. It is characterized by passing control from the main program to the corresponding module in the processing module group by calling a branch instruction table based on the value of the branch destination.

(Function) As described above, by providing a processing module group consisting of modules prepared for multiple processing contents, an index table indicating processing procedures, and a branch instruction table for branching to each processing module, non-standard data can be processed. Achieve efficient processing.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing a non-standard data processing program control method according to an embodiment of the present invention.

In the figure, 1 is the main program that receives processing requests from multiple terminals, infers processing content from human data,
An optimal table is selected from the index table group 3, and control is passed to each processing module according to the table contents. Reference numeral 2 denotes a processing module group, which performs a detailed check of the input data, and if the data is applicable, executes the desired process and outputs the data; if the data is not applicable, the processing returns to the main program 1. 3 is a group of index tables accessed by the main program 1, and "variables" are used to select a branch instruction table based on the tendency of input data in advance.
It consists of a collection of combinations of. A branch instruction table group 4 is accessed by the output from the index table group 3, and is made up of a collection of branch instructions to each processing module 3.

FIG. 2 shows a processing flowchart of the present invention.

Step S1 is a step of inferring from the keywords in the non-standard data and selecting one of the processing modules 2 to be executed next from the index table 3.

Step S2 is a step in which the human data is checked in detail, and if the result of the check is that the processing data is applicable, it is executed, and if the data is not applicable, the processing moves to the next processing module.

As mentioned above, by providing a processing module group 2 consisting of modules prepared for a plurality of processing contents, an index table group 3 indicating processing procedures, and a branch instruction table 4 for branching to each processing module, non-standard data can be processed. Achieve efficient processing.

In other words, keywords are searched from the data group input to the main program 1, and the contents of the data are inferred from the keywords.For example, if there are many keywords related to product quality, it is inferred that the data should be relayed to the quality manager.) , the main program 1 accesses the index table group 3 and calls the branch instruction table 4 based on the value of the branch destination in the table in the index table group 3.

As a result, control is passed from the main program 1 to the corresponding module in the processing module 11-rule group 2.

Further, by rearranging the variables in the index table group 3, the execution procedure of the processing module can be changed arbitrarily, and program control that can flexibly respond to the data content to be processed becomes possible.

Next, a specific example of control will be explained with reference to FIG.

In this example, the input data patterns in index table group 3 are pattern 1. Pattern 2. A case is shown in which the pattern 3 is written and the branch instructions in the branch instruction table group 4 are written as BASIC instructions. Also,
The broken line in FIG. 3 shows the flow of control.

Assume that as a result of rough inference of the input data using the main program 1, pattern 3 is selected from the index table group 3. Here, pattern 3 is set so that processing module B and processing module C of processing module group 2 are executed in order, and from this pattern 3, main program l branches to select processing module B from branch instruction table group 4. The command is executed and control is transferred to processing module B.

Then, if processing is possible in the processing module B, the processing is executed and then the main program 1 is returned to. If it is not applicable, no processing is performed and the program returns to the departmental program 1.

Further, if the input data has been processed, the main program 1 enters another input data receiving state, and if the input data has not been processed, it passes control to the processing module according to the index table group 3.

The present invention can be applied to an intelligent e-mail system using non-standard data, a 24-hour automatic data collection and distribution system, and the like.

[Effects of the Invention] As described above, according to the present invention, a module that is closer to the processing conditions is selected, so that the target processing can be effectively controlled.

Furthermore, in program maintenance, it is possible to respond to specification changes by maintaining the index table without modifying the program logic, resulting in excellent effects in maintaining program quality.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an overview of a non-standard data processing program control method according to an embodiment of the present invention, and FIG.
FIG. 3 is a flowchart showing the control of the non-standard data processing program control method shown in FIG. 1, and FIG. l...Main program, 2...Processing module group, 3
... Index table group, 4... Branch instruction table group.

Claims (1)

[Claims] In a computer system in which a plurality of processing modules process the contents of atypical data randomly generated from a plurality of terminal devices, etc., the main program infers the contents of the data from the keywords of the input data group. , accessing an index table that describes the optimal processing module as a result, and calling a branch instruction table based on the branch destination value in the index table, thereby passing control from the main program to the corresponding module in the processing module group. An atypical data processing program control method characterized by: